The long-term goal of this project is to integrate the analysis of all genomic studies of the GAML PPG to identify and validate acquired genetic changes that may contribute to the pathogenesis of AML. Somatic mutations that are predicted to change the function of a gene will be analyzed to assess their consequences on patient outcomes, on patterns of gene expression, and on AML-relevant pathways. To identify these mutations, high-resolution genomic screens and large scale resequencing studies will be required. In the GAML PPG, several cores and projects are generating genome-wide databases that must be carefully coanalyzed to identify candidate genes for resequencing and validation studies. These databases also must be mined to define relationships between the mutations and the biological pathways that they affect. To achieve these goals, we propose these Specific Aims: Specific Aim 1: We will analyze the outputs of multiple genomic screens to prioritize candidate genes for resequencing, and we will define and validate somatic mutations in AML samples. Exonbased resequencing studies (Core D) are difficult and expensive to perform, and candidate genes for these studies must therefore be carefully prioritized using data generated by high-resolution genomic screens. Array-based gene expression profiling, high resolution array-based comparative genomic hybridization, and high resolution array-based SNP genotyping studies will be used to identify genes and/or loci that are deleted, amplified, or duplicated from one parental allele (uniparental disomy), or that have aberrant patterns of expression;whole genome resequencing studies of 10 M1 AML genomes (Project 1) will also be analyzed to define potentially important mutations that will be validated using the 94 matched tumor-germline AML samples from the Discovery Set. When potentially important somatic mutations are identified, we will perform additional studies to verify the mutation and its frequency in the AML sample of interest (i.e. bacterial cloning and resequencing of the mutant exon in 96 clones from the sample), and we will further define the mutation's frequency in 94 fully annotated AML cases obtained from Cancer and Leukemia Group B (CALGB). Specific Aim 2: We will define the clinical and gene expression consequences of validated somatic mutations, and define biologic pathways altered by these mutations. We will use statistical approaches to define the effects of mutations on clinical outcomes. Novel informatics approaches (e.g. promoter analyses, pathway/interaction network construction, etc.) will be used to define the effects of mutations on patterns of gene expression, and to identify potentially relevant biological pathways that are affected by AML mutations. These algorithms will be used to identify additional genes for resequencing studies. Selected mutations and pathways identified by these studies will be biologically validated in the laboratories of PPG members.